Age-specific incidence curves from the state of Connecticut reveal a small peak in childhood followed by a much taller, sharper peak with a maximum between ages 55 and 65. This also holds true for data collected from most large population-based registries. In constrast to this are the age-specific incidence curves for Rochester, Minnesota, which show a steady increase in incidence with increase in age. In addition, the reported incidence rate at any particular age for Rochester, Minnesota, is higher than comparable rates from other registries. The reasons for these discrepancies are currently being analyzed. A manuscript concerning the findings has been prepared and has been accepted for publication.